A Phrase-Based Context-Dependent Joint Probability Model for Named Entity Translation

Zhang, Min; Li, Haizhou; Su, Jian; Setiawan, Hendra

doi:10.1007/11562214_53

Cited by 9 publications

(12 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are two kinds of transliteration-based statistical method. One is phoneme-based transliteration method [3,5,6]. Another is grapheme-based transliteration method [3,6].…”

Section: Statistical Transliteration Methodsmentioning

confidence: 99%

“…One is phoneme-based transliteration method [3,5,6]. Another is grapheme-based transliteration method [3,6]. There are three conversion steps for phoneme-based transliteration method: grapheme-to-phoneme conversion, phoneme-to-phoneme conversion and phoneme-to-grapheme.…”

Section: Statistical Transliteration Methodsmentioning

confidence: 99%

“…(1) Related work of parallel corpora-based method Fei Huang [2,8], Chen Hua-xing [9],Zhang min [6], Li [10] and Chen YuFeng focused on finding NE correspondence in parallel corpora [1]. Fei Huang's named entity alignment model incorporated named entity transliteration cost, word-based named entity translation cost and named entity tagging cost to align bilingual named entity equivalences.…”

Section: Parallel Corpora-based Methodsmentioning

confidence: 99%

“…Li Tingting used the conditional random field to recognize Japanese names, and kana names were translated using Moses machine translation system [10]. Zhang Min proposed a phrase-based context-dependent joint probability model for named entity translation [6]. Named entity translation is similar to phrase-based statistical machine translation: target phrases were generated by the lexical mapping model and word reordering was performed by the permutation model at the phrase level.…”

Section: Parallel Corpora-based Methodsmentioning

confidence: 99%

“…LDC2005T34 was used for the corrective alignment. Zhang Min [6] used LDC Chinese-English NE translation corpus. A bilingual corpus aligned in the source language order is used to train lexical mapping model, and a target language corpus with phrase segmentation in their original word order is used to train permutation model.…”

Section: (2) Parallel Corporamentioning

confidence: 99%

See 4 more Smart Citations

The Technical Analyses of Named Entity Translation

Liu¹

2015

Proceedings of the 2015 International Symposium on Computers &Amp; Informatics

View full text Add to dashboard Cite

There are three methods: rule-based method, statistical method and web mining method for named entity translation. The rule-based method did not achieve satisfactory results. High-quality translation equivalents can be obtained from parallel corpora for statistical method, and a prerequisite is the availability of a large scale of annotated corpora. The comparable corpora are easier to obtain than parallel corpora. But translation extraction from comparable corpora achieves lower accuracy than that of parallel corpora. Web mining method can acquire the translation of high-frequency named entities and it is difficult to translate the low-frequency named entities.

show abstract

“…There are two kinds of transliteration-based statistical method. One is phoneme-based transliteration method [3,5,6]. Another is grapheme-based transliteration method [3,6].…”

Section: Statistical Transliteration Methodsmentioning

confidence: 99%

Section: Statistical Transliteration Methodsmentioning

confidence: 99%

Section: Parallel Corpora-based Methodsmentioning

confidence: 99%

Section: Parallel Corpora-based Methodsmentioning

confidence: 99%

Section: (2) Parallel Corporamentioning

confidence: 99%

See 3 more Smart Citations

The Technical Analyses of Named Entity Translation

Liu¹

2015

Proceedings of the 2015 International Symposium on Computers &Amp; Informatics

View full text Add to dashboard Cite

show abstract

Named entity transliteration with sequence-to-sequence neural network

Chng

2017

2017 International Conference on Asian Language Processing (IALP)

View full text Add to dashboard Cite

A Structure-Based Model for Chinese Organization Name Translation

Chen

Zong

2008

ACM Transactions on Asian Language Information Processing

View full text Add to dashboard Cite

Named entity (NE) translation is a fundamental task in multilingual natural language processing. The performance of a machine translation system depends heavily on precise translation of the inclusive NEs. Furthermore, organization name (ON) is the most complex NE for translation among all the NEs. In this article, the structure formulation of ONs is investigated and a hierarchical structure-based ON translation model for Chinese-to-English translation system is presented. First, the model performs ON chunking; then both the translation of words within chunks and the process of chunk-reordering are achieved by synchronous context-free grammar (CFG). The CFG rules are extracted from bilingual ON pairs in a training program.The main contributions of this article are: (1) defining appropriate chunk-units for analyzing the internal structure of Chinese ONs; (2) making the chunk-based ON translation feasible and flexible via a hierarchical CFG derivation; and (3) proposing a training architecture to automatically learn the synchronous CFG for constructing ONs with chunk-units from aligned bilingual ON pairs. The experiments show that the proposed approach translates the Chinese ONs into English with an accuracy of 93.75% and significantly improves the performance of a baseline statistical machine translation (SMT) system.

show abstract

A Phrase-Based Context-Dependent Joint Probability Model for Named Entity Translation

Cited by 9 publications

References 20 publications

The Technical Analyses of Named Entity Translation

The Technical Analyses of Named Entity Translation

Named entity transliteration with sequence-to-sequence neural network

A Structure-Based Model for Chinese Organization Name Translation

Contact Info

Product

Resources

About